It has already been found that the round shape of holes can be changed into hexagonal shape during plasma etching processes. This work aims to understand the mechanism behind such a shape change using particle simulation method. The distribution of electric field produced by electrons was calculated for different heights from the mask surface. It is found that the field strength reaches its maximum around a hole edge and becomes the weakest between two holes. The field strength is weakened as moving away from the surface. The spatial distribution of this electric field shows obvious hexagonal shape around a hole edge at some distances from the surface. This charging distribution then affects the trajectories of ions that fall on a mask surface so that the round hole edge is etched to become a hexagonal hole edge. The changing of this hole shape will again alter the spatial distribution of electric field to enhance the charging effect dynamically.

The edge roughness (ER) developed in the mask pattern during the plasma etching process harms the perfect pattern transformation from mask to substrate. To understand and ultimately manipulate plasma-induced ER, this study investigated the interplay between charging and nanoscale roughness of an isolated rough mask hole in the plasma etching process using a modeling framework, which consisted of a surface etching module, a surface charging module, and a profile evolution module. Specifically, on the one hand, the distributions of the spatial electric field (E-field) and etching rate were simulated for the rough mask surface being etched. It is revealed that the distribution of the etching rate is similar with that of E-field clinging to the mask surface, and both of them reach their maximum strength around the mask hole edge and gradually becomes uniform/nonuniform with the decrease of the value of the dominant amplitude/wavelength of roughness; on the other hand, a string algorithm was used to simulate the evolution rule for a rough profile of a mask hole with etching time under various values of roughness parameters. The simulated evolution of the profile has good agreement with experimental observation. The charging effect contributes to the enhancement of the root mean square roughness. Additionally, the influence of roughness parameters on the charging time versus root mean square was also examined. The mechanism behind these results was analyzed systematically. This study will greatly contribute toward improving the physical and chemical properties of the mask or optimizing the etching technique.K E Y W O R D S charging, particle simulation, plasma etching, roughness

No abstract

The perfect pattern transfer from mask to substrate during the plasma‐etching process is strongly limited by the charging effect on the mask surface, which is increased by the accumulation of negative charges on the surface. These are mainly caused by high‐velocity isotropic electrons impinging on the mask surface faster than ions. These anisotropic ions thus bombard the undesired locations of the mask under the influence of the electric field (E‐field) established by electrons. This problem leads to significant damages to the mask pattern and causes deformations of etched features due to failure pattern transfer. This study examined that electron angular distribution (EAD; relative to the vertical direction, which can be regulated by voltage waveform tailoring) displays a close relationship with the mask pattern damage. Based on a modeling framework that consists of a surface etching module, a surface charging module, and a profile evolution module, the effects of changing the EAD on distributions of spatial E‐field and etching rate were studied focusing on an isolated rough mask hole surface. It is revealed that by narrowing the EAD shape, the E‐field strength and the etching rate around the mask hole edge can be reduced strongly, meanwhile, the number of electrons penetrating into the bottom of the trench can be greatly increased. These developments will supposedly reduce the mask‐pattern damage and improve the etching of high‐aspect‐ratio (HAR) features into the substrate. The simulated evolution rates of profile of a rough mask hole and the profile of E‐field strength inside the hole under various EAD shapes verify the above conclusions. The mechanism behind these results was analyzed systematically. This study provides a significant point for further investigation into the optimization of the etching technique.

No abstract